Types of database management system and their evolution

     In the past Database may be a single big table from which they retrieve all the knowledge. But today we've moved on from this simplistic definition of databases.

     The timeline varies from the 1980s to the present date and isn't exhaustive of all sorts of data management systems.

     Flat File Database is perhaps the simplest to know but at the present rarely used. this is often one huge table. Such sort of datasets was used back within the 1990s when data was only wont to retrieve information just in case of concerns. Very primitive analytics were possible on these databases.

     People start realizing that It brought a lot of redundant data at every entry. When Data repeats, People thought of storing this as different tables and define a hierarchy to access all the info, which can be called a hierarchical database.

     This is very almost like the folder structure on the laptop. Every folder can contain a sub-folder and that they can still hold more sub-folders. In some folders, we'll store files. However, every -the folder will have one parent Folder. Finally, we will create a hierarchy of the dataset. Its applications are restricted to one-to-one mapping data structures. Then People thought of database structures that will have different sorts of relations. this sort of structure should leave one-to-many mapping. Such a table came to be referred to as the electronic database management system.

     There are multiple keys that will help us merge different data sets during this database. this type of knowledge storage optimizes disk space occupied without compromising on data details. this is often the database that's employed by the analytics industry. However, when the info loses structure, such a database is going to be of no help.

     NoSQL is typically mentioned as “Not Only SQL”. Then people realized that unstructured text carries tonnes of data that they're unable to mine using RDBMS, anything which isn't RDBMS today is loosely referred to as NoSQL.

     If we attempt to store the whole data in RDBMS, for executing one, we'd like to hitch multiple tables with trillions of row together to seek out a combined table then run algorithms to seek out the foremost relevant information for the user. This doesn't look to be a second job. Hence we'd like to maneuver from a tabular understanding of knowledge to a more flow-based arrangement. this is often what brought NoSQL structures.

     Databases form the inspiration in the analytics industry. albeit we don’t know all of them intimately, we should always have a summary of the whole spectrum of databases.